What reasons make the use of multithreading impractical for performing computational tasks?

The reasons why multithreading may be impractical for computational tasks in Python are related to the features of the interpreter and the language architecture:

• GIL (Global Interpreter Lock) — in the standard implementation of Python (CPython), there is a global lock on the interpreter that prevents the execution of Python bytecode in multiple threads simultaneously. This limits the effectiveness of multithreading for CPU-bound tasks.

• Overhead of context switching — managing threads requires resources, and performance can degrade with a large number of threads.

• It's better to use multiprocessing — for compute-intensive tasks in Python, the multiprocessing module is usually used, which launches multiple processes, bypassing GIL limitations and allowing efficient use of multiple CPU cores.

Thus, multithreading in Python is more suitable for I/O-bound tasks, while for computations, multiprocessing is preferred.